Information Technology Reference
In-Depth Information
psychological make-up of the farmer, over characteristics of the farm household and
the structure of the farm business to the wider social milieu ([8]). However, little is
known about the relative contributions of these various factors in varying contexts and
a general model is lacking. In order to reduce the uncertainty of model outcomes with
respect to the choice of (farmers') decision-making rules, a robustness analysis that
compares model results under various assumptions would be ideal to substantiate the
conclusions drawn from agricultural land-use models. In most cases such an analysis is
not conducted due to time/budget constraints, although the studies cited above indicate
that this would be important.
In this paper we present and compare two empirically grounded agent-based
models of agricultural land-use change and associated groundwater over-use in the
Upper Guadiana, Spain. The models have been implemented to reconstruct the history
of agriculture in the Upper Guadiana Basin in order to identify characteristics of
farmers which are important to understand land-use change in that region. The models
use an identical general setting including options available to farmers (prices, costs
related to crop area planted, labour intensity, yields, water needs, etc.) and
regulations, and partly identical implementations of farmers with regard to the
diffusion of innovations and path-dependency arising from historical developments of
the farm. But they implement two different decision-making rules for farmers: one is
based on utility ([10]) and one is based on satisficing ([11]). We present the different
models and compare them with regard to similarities and differences of simulation
results in order to draw lessons on substantial and implementation-dependent results.
The next section introduces the case study. Section 3 then outlines the similarities
and differences between the two models and compares simulations to empirical data.
Section 4 then compares the models with respect to the factors driving the dynamics.
Section 5 draws the conclusions.
2
The Upper Guadiana Case Study
The Upper Guadiana Basin (UGB) is a rural area located in the Autonomous Region
Castilla La Mancha in central Spain. The UGB is one of Spain's driest areas ([12]),
the population density is very low and the UGB is classified as a rural area. Irrigation
of farm land accounts for approximately 90% of total groundwater use while
irrigation using surface water is hardly significant ([13]). During the last decades, the
amount of irrigated farming has increased and farming practices have changed
towards water-intensive crops, especially in the Mancha Occidental aquifer (MOA),
the area's main aquifer. In the MOA around 17.000 farms exist, and it accounts for
90% of the UGB's groundwater extractions ([14]). Irrigated agriculture encompasses
the possibility to plant water-intensive crops like high-yield cereals (maize, alfalfa)
and melons, which do not grow in the region without irrigation. It further provides the
possibility to achieve higher yields of traditional crops like winter wheat and barley as
well as of vineyards and olives. All this led to a considerable expansion of irrigated
agriculture in the UGB until the mid 1980s. The associated overexploitation of
groundwater resources in the MOA endangered wetlands of high ecological value
